Fig. 2. Simultaneous imaging & communication [J5]

Project 1: Foundations for Imaging with Wireless Networks

We aim to develop the foundational theoretical framework for imaging using wireless networks. Specifically, we address the following two inter-related questions: 

1. First, what are the fundamental limits on wireless imaging networks with arbitrary geometries and realistic propagation environments? 
2. Second, what are the fundamental tradeoffs in imaging and communications, when the same network aims to perform them both simultaneously?

We have made progress towards answering both questions in recent work:

1. In [J4], we tackle the first challenge by characterizing the impact of multipath on imaging resolution – see Fig. 1. Specifically, we use an information theoretic degrees of freedom framework to answer the question: When does multipath offer gains in imaging resolution over line-of-sight-only propagation? Our analysis unifies prior theoretical limits derived for line-of-sight propagation with numerical and experimental results with multipath.
2. In [J5], we tackle the second challenge by characterizing the fundamental performance trade-offs for joint imaging and communication performed simultaneously by the same network – see Fig. 2. We consider an uplink system configuration with a full-duplex base station illuminating an imaging scene (in monostatic mode) while receiving uplink data from a mobile user. Our main contributions are two-fold. First, we propose a unified signal space analysis framework based on the degrees of freedom metric to characterize imaging resolution-communication rate trade-offs in the high signal-to-noise ratio regime. Second, we propose a dual-function joint processing scheme, decode-and-image, that allows the BS to simultaneously form an image of the scene while decoding the uplink user’s data. Our analysis and proposed scheme highlight the benefits of exploiting the uplink signals for imaging, at the cost of increased cooperation between the BS and uplink user.
3. We have furthermore extended some of our results to tackle the challenging case of joint imaging & communication performed by a distributed network of nodes. In [C8], we extend the scheme from [J5] that allows multiple cooperating base stations to simultaneously decode an uplink user’s message and form an image of the surrounding environment. Our proposed scheme achieves similar imaging and communication performance as the centralized scheme proposed in [J5], with order-wise reduction in computational complexity.

Project 2: Orientation Estimation with Active Backscatter Tags

We considered the problem of estimating the orientation of a 3D object with the assistance of configurable backscatter tags. In [C2,C10,P1] explored the idea of designing tag response codes to improve the accuracy of orientation estimation, by connecting to ideas from error-correcting codes (but in real domain). Our initial exploration in [C2] resulted in promising numerical designs based on heuristic rules. We significantly improved our ideas in [C10,P1] where we proposed two principled code design criteria, for average and worst case criteria. We also developed a lower bound on the worst-case error (based on the Le Cam method), which quantifies the systems performance with respect to channel parameters such as the number of antennas and the number of tags used. Through comprehensive numerical exploration of the systems performance, including the robustness of the design criteria against imperfect channel knowledge, we showed that our design criteria yield codes that offer significant advantages over the channel-oblivious orthogonal code for both the average and worst-case error performance. Moreover we also found that precise channel knowledge at the transmitter is not critical; our design criteria are robust against channel estimation errors, retaining almost all improvements.